How to Separate Branded and Non-Branded Traffic in Search Console

Key takeaways

• Define the brand rule before comparing periods.
• Review ambiguous queries rather than forcing labels.
• Keep native and custom classifications clearly identified.

Separate branded and non-branded traffic in Google Search Console to see how known-brand demand differs from other search discovery. Use the built-in labels when it is available. If you need a custom rule, build and review a search filter. In both cases, write down the rule first. Then compare the results. The split is a way to read query data, not a ranking factor or a perfect measure of new customers.

Define what counts as a brand search for your business

Start with a written rule. Include the firm’s name, common spelling variants, and product names that clearly identify the firm. Decide how to handle names shared with ordinary words or other firms. A rule that no one can explain will produce a weak report. Precise numbers do not fix an unclear rule.

Keep an unclear group during the review. For example, a short brand name may also name a material or place. Do not force every unclear query into “brand” just to complete a chart. Record examples and ask the owner which meanings are relevant. Apply the same rule to both comparison periods.

Check the native branded-query filter first

In the Search results report, look for the branded and non-branded search filter. Google provides these labels for sites that meet its requirements. Sites with few impressions may not have the filter. Review the current report rather than promising that every account has the same controls.

Use a few known queries to assess the labeling. Check a clear brand name, a clear generic service search, and an unclear term. Google notes that some queries may be placed in the wrong group. Keep that limit beside the results. Do not silently treat the tool’s label as a verified fact about every searcher.

Build a fallback query rule when you need one

If the built-in filter is absent, use a search rule. You can also use a rule when the built-in labels do not fit your needs. Begin with the simplest rule that captures your brand variants. A single name may need only a contains filter. Several names may need a regular expression. This is a text pattern used to match queries. Test the expression on known examples before you use the totals.

Write the expression in plain English beside its technical form. A fictional business called Cedar Lantern might include its full name and a common joined spelling. It should not by default include every query containing “cedar.” That broad rule would label many generic searches as brand demand.

Review inclusions and exclusions with a small test set

Build a test list with examples that should match, examples that should not, and examples that need a decision. Run the rule against that list. Check punctuation, spacing, and misspellings that occur in the visible queries. Avoid adding every imaginable typo when there is no evidence that it matters.

Review the non-branded side too. A rule that captures the exact company name may still leave product-brand queries outside it. If you change the rule, give it a new version and rerun both periods. A narrow rule last month and a wide rule this month can create a false growth story. Use the same rule for both.

Use the same scope for the two groups

Hold the property, search type, dates, country, device, and page scope steady. Change only the brand condition when making the split. Label exports so another person can repeat the view. A comparison is useful only when the reader knows what belongs in each group.

Keep any data that is missing or has no label visible in your explanation. Query-level report has limits, so do not force the two filtered totals to reconcile perfectly with an unfiltered headline. A neat percentage chart should not hide gaps in what the report shows. State the scope of the split you actually calculated.

Interpret brand and non-brand results without assigning motives

A branded query shows a name in the search. It does not prove the person is an existing customer. A first-time buyer may have heard of the company from a friend or an ad. A generic query may come from someone who already knows the firm. Use the labels to group searches. They do not identify who the customers are.

Avoid one healthy ratio for all sites. A new specialist service, a known retailer, and a business running a large brand campaign can have very different mixes. Compare each group with its own history and the firm’s context. Ask what changed before deciding that one group should be larger.

Example: a campaign changes the total picture

Imagine a fictional lighting business that runs a local brand campaign. Branded clicks rise while non-branded clicks stay flat. Total search clicks increase. The split shows where the change occurred, but it does not prove that the campaign caused every added click. Other events may have happened in the same period.

The team records the campaign dates and checks the brand-query group. It keeps its generic service pages under a separate review. It does not report the overall increase as proof that those pages rank better for new searches. The segmentation helps the team ask a more precise question.

Put the split into the DEX review workflow

DEX connects to Search Console and organizes core search metrics for content decisions. Use those trends to decide where to investigate. Use the source report for brand labeling and custom filtering unless you have verified an matching control in your DEX account. Do not assume that a connected metric includes every source-tool feature.

Keep the meaning tied to the products and services you sell. Generic discovery matters when it brings relevant searches, not merely because it makes the non-brand share look larger. Brand demand matters too. The useful question is which pages and offers need the next action. If a group earns many impressions but few visits, investigate low CTR within each group before changing its page titles.

Test the rule with real query examples

For the fictional Cedar Lantern business, put three searches in a test sheet. “Cedar Lantern lamp repair” should match the brand rule. “Cedar fence repair” should not. “Cedar Lantern reviews” should match, but its task differs from a quote request. These short tests check both the rule and how you read the result.

Add a row for a name that could mean two things. Mark it for review. Ask the owner what the name means in that search before you add it to the rule. Keep a note of the choice. Run this small test set each time you change the rule so a new fix does not break an old match.

Keep a reusable classification note

Save the rule, version date, owner, sample queries, exclusions, and comparison scope. State whether the figures came from Google’s native labels or your own filter. Do not switch methods silently between reports. If the firm renames a product, review the rule. Check whether you can still fairly compare the old and new data.

Finish with an explicit finding and a limit. For example: brand-query clicks rose within the fixed report scope, while generic-query clicks were stable. The note should also state any unclear labels. That is enough to support a focused next review without turning a search filter into a claim about customer growth.

FAQ

Does branded mean an existing customer?

+

No. A first-time visitor may search for a name they just heard.

Can I use a custom brand definition?

+

Yes. Test the query rule and use it consistently across periods.

Is there a healthy brand share for every site?

+

No. Interpret each group in its business context.

Sources

1. Official guidance · Google
2. Performance report overview · Google

Dex SEO

Dex SEO is an AI SEO agent for small businesses. Every article here is planned from search data, drafted by Dex, fact-checked against the sources it cites, and reviewed by the team before it is published.

How this article was made: planned and drafted by Dex from search data, fact-checked by Dex (5 claims checked), then approved by the Dex SEO team.